[论文解读] A Comparison Study of Credit Card Fraud Detection: Supervised versus Unsupervised

Xuetong Niu, Li Wang+1 more|arXiv (Cornell University)|Apr 24, 2019

Imbalanced Data Classification Techniques参考文献 22被引用 72

一句话总结

本文在 Kaggle 数据集上比较了 6 个有监督与 4 个无监督的信用卡欺诈检测模型，使用 AUROC 与 5 折交叉验证，结果表明有监督模型总体略优于无监督模型。

ABSTRACT

Credit card has become popular mode of payment for both online and offline purchase, which leads to increasing daily fraud transactions. An Efficient fraud detection methodology is therefore essential to maintain the reliability of the payment system. In this study, we perform a comparison study of credit card fraud detection by using various supervised and unsupervised approaches. Specifically, 6 supervised classification models, i.e., Logistic Regression (LR), K-Nearest Neighbors (KNN), Support Vector Machines (SVM), Decision Tree (DT), Random Forest (RF), Extreme Gradient Boosting (XGB), as well as 4 unsupervised anomaly detection models, i.e., One-Class SVM (OCSVM), Auto-Encoder (AE), Restricted Boltzmann Machine (RBM), and Generative Adversarial Networks (GAN), are explored in this study. We train all these models on a public credit card transaction dataset from Kaggle website, which contains 492 frauds out of 284,807 transactions. The labels of the transactions are used for supervised learning models only. The performance of each model is evaluated through 5-fold cross validation in terms of Area Under the Receiver Operating Curves (AUROC). Within supervised approaches, XGB and RF obtain the best performance with AUROC = 0.989 and AUROC = 0.988, respectively. While for unsupervised approaches, RBM achieves the best performance with AUROC = 0.961, followed by GAN with AUROC = 0.954. The experimental results show that supervised models perform slightly better than unsupervised models in this study. Anyway, unsupervised approaches are still promising for credit card fraud transaction detection due to the insufficient annotation and the data imbalance issue in real-world applications.

研究动机与目标

• 评估并比较有监督与无监督的机器学习模型在信用卡欺诈检测中的性能。
• 评估数据标注、类别不平衡和注释延迟如何影响模型性能。
• 确定在真实欺诈数据集上，哪些模型族（有监督 vs 无监督）能够获得更高的 AUROC。

提出的方法

• 使用带标签的欺诈数据并通过下采样将类别平衡来评估 6 种有监督模型（LR、KNN、SVM、DT、RF、XGB）。
• 在无标签数据上训练的 4 种无监督模型（OCSVM、AE、RBM、GAN）用于检测异常。
• 采用 5 折交叉验证，并以 AUROC 作为性能度量。
• 用 RobustScaler 对 Time 和 Amount 进行归一化，并将非欺诈样本下采样，使欺诈样本与之相等（各 492 个）。
• 在交叉验证内使用网格搜索对超参数进行调参。
• 提供所使用的 AE 与 GAN 架构的实现细节。

实验结果

研究问题

• RQ1 在一个真实且高度不平衡的数据集上，有监督模型在 AUROC 上与无监督模型在信用卡欺诈检测中的表现有何差异？
• RQ2 在本研究中，每个类别（有监督 vs 无监督）中的表现最佳模型是什么？
• RQ3 在标注要求和数据不平衡的前提下，有监督与无监督方法之间的实际权衡是什么？

主要发现

模型	AUROC
XGB (Supervised)	0.989–0.990
RF (Supervised)	0.988
DT (Supervised)	0.95
LR (Supervised)	not specified in abstract
KNN (Supervised)	not specified in abstract
SVM (Supervised)	not specified in abstract
RBM (Unsupervised)	0.961
GAN (Unsupervised)	0.954
AE (Unsupervised)	not specified in abstract
OC-SVM (Unsupervised)	0.90

• XGBoost（有监督）在该数据集上实现了最高的 AUROC，0.989–0.990。
• Random Forest（有监督）实现 AUROC 为 0.988。
• Decision Tree（有监督）实现 AUROC 为 0.95（在有监督模型中最低）。
• RBM（无监督）实现 AUROC 为 0.961，使其成为本研究中无监督方法的最佳者。
• GAN（无监督）实现 AUROC 为 0.954。
• OC-SVM（无监督）实现 AUROC 为 0.90，是无监督模型中的最低值。